Mother tongue may influence musical ear - speech characteristics influence how people hear music

Science News, Dec 1, 1990 by Robert N. Langreth

Mother tongue may influence musical ear

When a Briton and a Californian listen to Beethoven's Fifth Symphony, they may not hear the same thing. New research indicates that people who speak different dialects of a language perceive tonal patterns in strikingly different ways, supporting long-standing speculations that speech characteristics influence the way people hear music.

Diana Deutsch of the University of California, San Diego, set out to investigate a musical paradox she had discovered four years ago (SN: 12/20 & 27/86, p.391). This phenomenon, called the tritone paradox, occurred when she electronically removed specific "overtones" from a series of two computer-generated pitches separated by a half-octave -- an interval called a tritone. Overtones normally help listeners identify the octave of a note; without them, the octave becomes ambiguous, Deutsch says.

In her 1986 study, some people who listened to paradoxical tritone series perceived a rise in pitch from one note to the next, while others perceived a descent. Also, whether an individual thought the second tone was higher or lower depended on the first note. For instance, a person might typically insist that C followed by F sharp was rising but that D followed by G sharp was descending.

Deutsch concluded that most people mentally arrange musical pitches on a circular map, or "pitch-class circle," placing all the notes in positions comparable to the numbers on a clock. However, she observed, the orientation of notes along the circle seems to vary from one individual to the next -- for instance, one person might position a particular note at the 12:00 site while another person places it at 5:00. This, according to Deutsch, determines how an individual perceives a paradoxical effect.

But a key question remained: Why do different people orient their mental maps in different ways?

Deutsch believes the answer lies in language. In the new study, she played several ambiguous tritone series to 24 people raised in southern California and then to 12 people raised in southern England. On average, she found, when the Californians thought a particular series rose, the English thought it descended -- and vice versa.

The results provide the first demonstration that listeners with different dialects differ profoundly in their perceptions of musical patterns, Deutsch asserted this week at a meeting of the Acoustical Society of America in San Diego. "A relationship between language and music has been hypothesized since ancient times," she told SCIENCE NEWS. "But what hasn't been known is if there is some direct influence of language on music."

"I've concluded very firmly that how you orient this circle of note names . . . is acquired by exposure to speech around you," she adds. Deutsch says her analyses appear to rule out many other potential causes of the tritone paradox, such as differences in age, gender, musical training or mechanics of hearing.

Nor does she believe genetic factors can explain the variation. "It doesn't seem reasonable [to conclude] that we're dealing with two genetic pools [in the latest experiment] if you consider that the population of the Californians was so heterogeneous genetically, including Asian Americans, German Americans" and other ethnic groups. The English group was also fairly heterogeneous, she says.

Deutsch suggests that paradoxical effects might also be heard in real music, perhaps in large orchestras. If so, language-linked perceptions "could drive a person's emotional response," she contends. "An audience in London would want one performance, but an audience in Los Angeles would want something different."

Bruno H. Repp, a linguist at the Haskins Laboratories in New Haven, Conn., remains skeptical. "Whether such effects can be heard in real music, I don't know. But I doubt it," he told SCIENCE NEWS.

At the same meeting, Deutsch reported on another study indicating that the fundamental frequency of a person's voice -- the lowest frequency present in a person's voice -- also affects how that person perceives the tritone paradox. She measured the fundamental frequencies of nine men and women, most of them college students in San Diego, and then tested their perception of the tritone paradox. In eight of the nine, she discovered a correlation between fundamental voice frequency and how the person perceived various tritone series.

The relationship between Deutsch's dialect findings and the voice-pitch correlations is perplexing. "I don't see why a cultural factor [such as dialect] would have any influence [on the perception of the tritone paradox]," Repp argues. On the other hand, he says, the results "would make sense . . . if different cultural groups used different fundamental ranges in speaking."

Deutsch did not measure the fundamental frequencies of the English/Californian volunteers, but she speculates that further studies may reveal voice-pitch differences between the two groups. She also plans to investigate how people who speak languages other than English react to the tritone paradox, and to compare people raised in various U.S. regions. Ultimately, Deutsch wants to see whether a live orchestra can produce the tritone paradox.